Subaru Impreza GC8 & RS Forum & Community banner
21 - 40 of 80 Posts

·
Registered
'98 OBS
Joined
·
648 Posts
I'm using my '98 ej22 block with some '98 ej25d heads for my td04(and hopefully vf34 later on) build. I'm de-shrouding the combustion chambers a bit and will use ej222 gaskets, hoping for right around 9:1 compression.

Good info in this thread, thanks for sharing
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #23 · (Edited)
Just to add...

EJ25D Cam Specs:
Intake opens 6° BDTC.
Intake closes 50° ABDC.
Exhaust opens 54°/30° BBDC.
Exhaust closes 10°/10° ATDC.
Overlap 16°.

EJ20G Cam Specs:
Intake opens 8° BDTC.
Intake closes 52° ABDC.
Exhaust opens 52° BBDC.
Exhaust closes 8° ATDC.
Overlap 16°.

EJ20K STi Cam Specs:
Intake opens 10° BDTC
Intake closes 56° ABDC
Exhaust opens 56° BBDC
Exhaust closes 12° ATDC
Overlap 22°

The 25D cams and the 20K cams are like night and day different. The 25D cams die off approximately at 6500rpm, and they start to die off around 5800-6000rpm. The 20K cams carry all the way past 6500rpm, and die off at 7500rpm. The 20K ECU is limited to 8200rpm. The 20K cams are hotter all around.

Now, 25D cams are pretty much N/A mirrors of late EJ20G cams, the engines being massively identical in the head department. The EJ20G is NOT the EJ20K, even though late variants look the same. It is a far more tame, more civil version that was found in non-STi turbocharged cars.

It was WAY more common that the STi stuff. I call them "Common G's" because back when these were ordered into the USA by the truckload, there were far more of these than anything. They're also more reliable than the 20K because of their lack of aggression.

Another thing to note is that some of the Common G cams and the EJ20H cams ARE THE SAME, as are the EJ20K cams to the EJ20R cams, as EJ20H heads are Common G heads and EJ20R heads are EJ20K heads, both drilled and tapped for twin turbocharging.
 

·
Experimenter
Tubaru Pickup
Joined
·
3,914 Posts
good for me stock ecu quits at 6200, i dunno how much more hp i coulda got with a higher rev limiter but it was still climbin



that odd graph was a 4th gear pull
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #25 ·
good for me stock ecu quits at 6200, i dunno how much more hp i coulda got with a higher rev limiter but it was still climbin

Get a 25D ecu and wire that up to whatever you're using. You will have a 6500rpm rev limit and a bit more playroom, but after 6500, you're cooked unless you upgrade the cams.
 

·
needs more cowbell
95L, 07 2.5i, R6
Joined
·
9,315 Posts
None taken, but for you to say its "common knowledge" you have to imply there is something different about my setup and the others who haven't failed. Metal is metal, and the ej20k blocks are built no worse than the previous g versions. Spinning a bearing or throwing a rod isn't always indicative of tune, it has also to do with care. They (ei20k) were tuned for more power with different ecus and different turbos and people push them as hard as they can until something breaks. The fact I and others don't share that curse is a testament its more to do with the way they are maintained. There is nothing different about the metal in my ej20k and the metal in the ones that have failed. You have to realize people may unintentionally run fuel that is bad or not actually 93 octane and may run worn out oil and/or low oil levels. When you are making around 300hp with a 2.0 engine and running those levels of boost, fuel and oil quality and quantity is critical. Other things like aging hoses and gaskets need to be addressed before even running the engine. I can't count the number of times I've seen people posting their setups (imagine the ones you don't see) being "done" and still having the original hoses and gaskets on 90's engines...expecting them to last. I started a thread to try and get count of how many engines are still going but regret it now that I have no way of actually knowing the ratio of 92-96 engines sold vs 97-98 (this ratio would indicate numbers would be skewed in the ej20g favor) and how many actually do what they should when swapping. There is no way to actually know without seeing every individual case and considering the conditions, especially knowing the ej20g setups came from the factory with substantially less power in the first place. Engine management improved in 99 but that isn't indicative of poor design of previous versions, only that subaru is constantly striving to make them even better.

The ej20"k" block would last just as long with an ecu designated for use with an ej20"g" block, that ecu just isn't pushing the envelope as much and there is a margin of safety for the people who attempt to abuse the setup. The block in itself is in no way inferior, in fact the heads are obviously superior overall. Don't throw in the rare closed blocks, having a closed block has nothing to do with spun bearing prevention.

Please explain what makes an ej22 case superior to an ej20 other than displacement being higher, making it easier to produce power.

My ej257 with k heads made 320wtq / 318hp with a 100mm bored stock liner case on a dyno dynamics back when tuned. It made 367wtq recently on a portable dyno at tx2k15. I run only 93 octane pump fuel and it is my daily driven work and fun car. I drove it to tx2k15 and actually ran the 1/4 for fun. Is my case also inferior to the ej22? If so, please explain.


Don't get me wrong, I appreciate you posting cam specs etc. I just don't understand how you come to your conclusions about "k" engines.

Do you also have the cam specs for ej20g sti models and ej20k non sti models? I notice you only posted ej20g non sti and ej20k sti.


No offense, but that's YOU and maybe a few others, but every single EJ20K or EJ20R I have ever seen run on pump gas has crapped out a rod bearing or a piston. Detonation is what does them in. If what I'm saying about the hybrids is a true thing, you can be sure that I'm speaking the truth about the JDM stuff as well, considering I've also torn into plenty of them. Let's not even talk about the drama shows the 9:1 EJ20's are. It's mind boggling. And don't say that there are EJ20R's and H's out there running on 93 as is and hitting boost daily. This is all factual stuff that is common knowledge.
 

·
Experimenter
Tubaru Pickup
Joined
·
3,914 Posts
the K's had issues in Europe until the point where they said no more mobile 1. I doubt it was a block issue, maybe subaru just subd the bearings out to a different supplier for those motors. The usdm ej205 is just as prone to bearing failure but lets face it. the majority of americans are terrible with maintanence

ivlike the 2.2 for its bigger bore on the less strenuous 75mm stroke.
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #28 · (Edited)
None taken, but for you to say its "common knowledge" you have to imply there is something different about my setup and the others who haven't failed. Metal is metal, and the ej20k blocks are built no worse than the previous g versions. Spinning a bearing or throwing a rod isn't always indicative of tune, it has also to do with care.
That's quite the opposite, as EJ20K's have open decks and weaker piston and rod bearing material than early EJ20G's. They're literally far removed from each other in build quality. That being said, a lot of the old 20G's I've seen torn open here weren't even close to breaking apart and had little wear on the rod bearings at 60k miles, as opposed to the 20K's, which had nearly 3 to 4x the wear.

Now, when the wrong fuel is used in an already out-of-shape engine, the wear accelerates because the internals just can't keep their integrity with the aggressive levels of detonation occurring in the chambers, which will eventually lead to blown rods, because the ECU's simply can't keep up with our octane. That being said, the Japanese youth of that time period often ran their engines hard without any concern, since they have to give up the car at a certain mileage anyway (emissions).

When those engines came to America and were run on already crappy-for-their-ECU's fuel, rods and pistons popped all of the place. It was like 20K's were puking their guts out. However, a lot of us found out that the reason why the old G's never did so was because they were simply too physically strong, and the reason why the Common G 8:1 never did so was because their programming was just tame enough.

This same rule applies with the EJ20R's. They too, were just about half dead off the boats.

They (ei20k) were tuned for more power with different ecus and different turbos and people push them as hard as they can until something breaks. The fact I and others don't share that curse is a testament its more to do with the way they are maintained. There is nothing different about the metal in my ej20k and the metal in the ones that have failed.
You're correct. You may have simply gotten one that was in prime condition. The majority weren't lucky.

You have to realize people may unintentionally run fuel that is bad or not actually 93 octane and may run worn out oil and/or low oil levels. When you are making around 300hp with a 2.0 engine and running those levels of boost, fuel and oil quality and quantity is critical.
This was already taken into account with the builders within the past 12yrs or do. Owners would do a full Stage Zero and use the correct fuel, and the 20K's still popped. It wasn't under people started breaking them down and examining the failure and ECU's that they started to realize what they were dealing with. Like you said, these situations were "critical", and that word was often used back them to describe how one should care for an EJ20K is they are to run one just out the crate.

Other things like aging hoses and gaskets need to be addressed before even running the engine. I can't count the number of times I've seen people posting their setups (imagine the ones you don't see) being "done" and still having the original hoses and gaskets on 90's engines...expecting them to last. I started a thread to try and get count of how many engines are still going but regret it now that I have no way of actually knowing the ratio of 92-96 engines sold vs 97-98 (this ratio would indicate numbers would be skewed in the ej20g favor) and how many actually do what they should when swapping. There is no way to actually know without seeing every individual case and considering the conditions, especially knowing the ej20g setups came from the factory with substantially less power in the first place. Engine management improved in 99 but that isn't indicative of poor design of previous versions, only that subaru is constantly striving to make them even better.
People started to order the JDM stuff once the new millenium hit, and during that time, they were learning the full potential of these engines, swapping, hybrid building, the whole 9. Now, while it is true that the early G's came with less power, they were also built with greater quality, which led to their longevity. Builders would extract 300whp at the closed deck 20G's and they would tank that power without the slightest bit in accelerated wear.

Enter the EJ20K/R. These were already high strung out the factory with weaker internals, which were already run halfway to death by the Japanese. Also, keep in mind that Japan is colder for longer periods of time than what we experience in the USA, and their fuel is substantially higher in octane for their performance cars to make up for that heavy air density.

The EJ20K/R ECU's came maxed out from STi, as aggressive as they could possibly get them to be (which means the Gentleman's Agreement was crap, because these 276bhp engines were actually more like 310bhp engines when the limiters were removed. 34bhp is a nice little chunk of power. And this is all from extremely primitive programming, since STi pretty much wrung out the K/R's as high as they could go with what they had at the time.

Now, for maintenance, I'm with you on that. I also encourage everyone to perform a Stage Zero (nomenclature taken from my Volvo days), in which case, they should replace and and ALL known worn supportive or direct parts, like air filters, radiator hoses, coils, spark plugs, fuel filters, etc. Not doing this before you run an engine is begging for trouble in the near future.

The ej20"k" block would last just as long with an ecu designated for use with an ej20"g" block, that ecu just isn't pushing the envelope as much and there is a margin of safety for the people who attempt to abuse the setup. The block in itself is in no way inferior, in fact the heads are obviously superior overall. Don't throw in the rare closed blocks, having a closed block has nothing to do with spun bearing prevention.
As I've mentioned before, the build quality between the EJ20K/R and the old EJ20G are quite different. Even if the G had a less aggressive ECU, it still powers an engine with tougher internals, so the longevity is extra compared to the highly sprung K/R's with their overly aggressive ECU's. The programming between the 2 are vastly different. If an old school G ecu ran a K, the K will last longer because it isn't asked to do much in comparison to it's own ECU, and this is before the mishaps of using 93 or lesser in them.

Also, every EJ20G from 1990 to late 1992 are tame variants that possess the closed deck block, so though they are hard to come by, they're not as rare as people may think. The late open deck G's, or Common G's, are built with similar quality to the 20K's, but they have much tamer ECU's, so they naturally don't get pushed as hard by their drivers.

If you run into one with a 9:1 compression ratio (despite what the general public says, they DO exist), do NOT run that engine on ANYTHING without replacing the block, the pistons, or tuning it. Those particular Common G's have TD04's and are bolted to automatic transmissions. Subaru did this to keep the power and efficiency balanced, but also tamed, since they believed the autos couldn't handle greater power.

Please explain what makes an ej22 case superior to an ej20 other than displacement being higher, making it easier to produce power.
It is the midway point between the EJ20 and EJ25, so low end torque is greater earlier in the rpm range, the EJ20 revolutions are retained (it uses an EJ20 crank), and you can bolt various EJ20 and EJ25 heads on it, maximizing potential than could be had from the EJ20's. One can also install an EJ25 crank into the case and maximize low-end torque and displacement at 2.35L, provided that you balance it to get rid of the upper rpm vibrations.

My ej257 with k heads made 320wtq / 318hp with a 100mm bored stock liner case on a dyno dynamics back when tuned. It made 367wtq recently on a portable dyno at tx2k15. I run only 93 octane pump fuel and it is my daily driven work and fun car. I drove it to tx2k15 and actually ran the 1/4 for fun. Is my case also inferior to the ej22? If so, please explain.
The EJ22 will always be inferior in potential when it comes to making power compared to the EJ25. That's not even a question. It's a simple matter of physics. There is no replacement for displacement. However, let's break down the mechanics to the EJ22 block, starting with the thicker liners. People have gone up to 27psi on these without so much as a cylinder wave or crack. The iron liners also hold up well in stock form.

The pistons are the next in line. Though not forged, they are strong enough that they can withstand 23-25psi with the correct tuning, air density, fueling, turbo, and heads. Many people have extracted as much as 380whp out of a stock EJ22E block before they lost ringlands. That being said, it's time to move onto the rods and crank, both of which are forged from the factory.

These have held up to 380whp without even a light bend. I've never seen a broken EJ22 crank, but I have seen broken EJ22 rods from lack of maintenance. Efficiency wise, since the crank and pistons are smaller, shorter stroked, and lighter than those in the EJ25, the efficiency as a hybrid for the EJ22 is greater, since there isn't as much internal inertia/resistance to spin up.

This all being said, the EJ25 in general is both inferior and superior to the EJ22 in many different ways. Build them up and sleeve them, and the EJ25 takes home the gold. But as a wild street engine, the EJ22 destroys it in reliability. It's a literal legend because all known EJ22 variants are known to be strong, as they can generally withstand more than the EJ25.

They were phased out because Subaru had to comply with emissions regulations here, and the EJ22 just couldn't keep up with the demands, just barely being able to qualify. It is an amazing engine, with lethargic cylinder heads that much be replaced with something possessing a head cc of greater than 44 if one is to begin to see an example of the potential this engine is capable of unlocking.

Don't get me wrong, I appreciate you posting cam specs etc. I just don't understand how you come to your conclusions about "k" engines.
Combination of years of notes, experience, and a LOT of financial mistakes for the sake of information gathering on these EJ's for the EJ22 hybrid builds.

Do you also have the cam specs for ej20g sti models and ej20k non sti models? I notice you only posted ej20g non sti and ej20k sti.
EJ20H 9:1 Cam Specs:
Intake opens 8° BDTC.
Intake closes 52° ABDC.
Exhaust opens 52° BBDC.
Exhaust closes 8° ATDC.
Overlap 16°.

EJ20H 8.5:1 Cam Specs:
Intake opens 10° BDTC.
Intake closes 50° ABDC.
Exhaust opens 52° BBDC.
Exhaust closes 2° ATDC.
Overlap 12°.

EJ20G STi Cam Specs:
Intake opens 6° BDTC.
Intake closes 56° ABDC.
Exhaust opens 55° BBDC.
Exhaust closes 11° ATDC.
Overlap 17°.
 

·
needs more cowbell
95L, 07 2.5i, R6
Joined
·
9,315 Posts
That's quite the opposite, as EJ20K's have open decks and weaker piston and rod bearing material than early EJ20G's. They're literally far removed from each other in build quality. That being said, a lot of the old 20G's I've seen torn open here weren't even close to breaking apart and had little wear on the rod bearings at 60k miles, as opposed to the 20K's, which had nearly 3 to 4x the wear.

The block I received had very little wear, I split it and replaced bearings and o rings anyway to verify internal condition. You can't say wear you've seen is due to poor design if you don't know history of the engine. How are you determining "nearly 3-4x" wear?

Now, when the wrong fuel is used in an already out-of-shape engine, the wear accelerates because the internals just can't keep their integrity with the aggressive levels of detonation occurring in the chambers, which will eventually lead to blown rods, because the ECU's simply can't keep up with our octane. That being said, the Japanese youth of that time period often ran their engines hard without any concern, since they have to give up the car at a certain mileage anyway (emissions).

"Our octane" Here it comes again. Please explain how US 93 octane is inferior to the Japanese premium. I've seen this before and watched it get shot down in a few threads.

When those engines came to America and were run on already crappy-for-their-ECU's fuel, rods and pistons popped all of the place. It was like 20K's were puking their guts out. However, a lot of us found out that the reason why the old G's never did so was because they were simply too physically strong, and the reason why the Common G 8:1 never did so was because their programming was just tame enough.

This same rule applies with the EJ20R's. They too, were just about half dead off the boats.

So now you're saying the ks are puking their guts out because of fuel and the gs didn't because they were just too strong and because of programming as well in the same paragraph. Please explain how strength of internals that you claim has anything to do with spun bearings. Even the closed deck versions are not going to have a better chance at preventing spun bearings.



You're correct. You may have simply gotten one that was in prime condition. The majority weren't lucky.

So now luck is why mine and many others' have made it out by the skin of their teeth? Look at my thread about the engines dying and you will notice that more than 50% of the ks show alive to the number of gs... and this is with the gs being produced for more than twice the number of years. This would indicate (if numbers were exact, but they can't be because not a large enough sample has been taken...but still) that the ks are as reliable or possibly better.

This was already taken into account with the builders within the past 12yrs or do. Owners would do a full Stage Zero and use the correct fuel, and the 20K's still popped. It wasn't under people started breaking them down and examining the failure and ECU's that they started to realize what they were dealing with. Like you said, these situations were "critical", and that word was often used back them to describe how one should care for an EJ20K is they are to run one just out the crate.

How are you figuring in all of the ecu variants used with the "k" longblocks? I ran the 1S and my engine came with a VF22 from the factory which even with the "tame" ecu the end product was not. I can assure you my engine made over the gentleman's agreement the entire time I owned it and is still in awesome shape "internally" as I've split the block out of curiosity.



People started to order the JDM stuff once the new millenium hit, and during that time, they were learning the full potential of these engines, swapping, hybrid building, the whole 9. Now, while it is true that the early G's came with less power, they were also built with greater quality, which led to their longevity. Builders would extract 300whp at the closed deck 20G's and they would tank that power without the slightest bit in accelerated wear.

There were obviously way more gs produced which contributes to their in service average being higher, and the fact they were originally tuned with less power contributes to their reliability overall. I sincerely doubt if you put an open deck g with an open deck k and tune them to the same hp levels and maintain them both equally, that one will be any more or less reliable. I've never seen any solid proof of this and doubt there ever will be, just a lot of conjecture and vague theory.

Enter the EJ20K/R. These were already high strung out the factory with weaker internals, which were already run halfway to death by the Japanese. Also, keep in mind that Japan is colder for longer periods of time than what we experience in the USA, and their fuel is substantially higher in octane for their performance cars to make up for that heavy air density.

Again with the "weaker internals", please explain how the g crank, rods, and pistons were stronger than the k? I'm talking about the non forged piston versions of course, to make things easier.

The EJ20K/R ECU's came maxed out from STi, as aggressive as they could possibly get them to be (which means the Gentleman's Agreement was crap, because these 276bhp engines were actually more like 310bhp engines when the limiters were removed. 34bhp is a nice little chunk of power. And this is all from extremely primitive programming, since STi pretty much wrung out the K/R's as high as they could go with what they had at the time.

Now, for maintenance, I'm with you on that. I also encourage everyone to perform a Stage Zero (nomenclature taken from my Volvo days), in which case, they should replace and and ALL known worn supportive or direct parts, like air filters, radiator hoses, coils, spark plugs, fuel filters, etc. Not doing this before you run an engine is begging for trouble in the near future.



As I've mentioned before, the build quality between the EJ20K/R and the old EJ20G are quite different. Even if the G had a less aggressive ECU, it still powers an engine with tougher internals, so the longevity is extra compared to the highly sprung K/R's with their overly aggressive ECU's. The programming between the 2 are vastly different. If an old school G ecu ran a K, the K will last longer because it isn't asked to do much in comparison to it's own ECU, and this is before the mishaps of using 93 or lesser in them.

Please tell me again how the g internals are "tougher", I've never seen proof of this.

Also, every EJ20G from 1990 to late 1992 are tame variants that possess the closed deck block, so though they are hard to come by, they're not as rare as people may think. The late open deck G's, or Common G's, are built with similar quality to the 20K's, but they have much tamer ECU's, so they naturally don't get pushed as hard by their drivers.

I"m not talking about the more rare closed deck versions, only the "normal" gs and ks.

If you run into one with a 9:1 compression ratio (despite what the general public says, they DO exist), do NOT run that engine on ANYTHING without replacing the block, the pistons, or tuning it. Those particular Common G's have TD04's and are bolted to automatic transmissions. Subaru did this to keep the power and efficiency balanced, but also tamed, since they believed the autos couldn't handle greater power.




It is the midway point between the EJ20 and EJ25, so low end torque is greater earlier in the rpm range, the EJ20 revolutions are retained (it uses an EJ20 crank), and you can bolt various EJ20 and EJ25 heads on it, maximizing potential than could be had from the EJ20's. One can also install an EJ25 crank into the case and maximize low-end torque and displacement at 2.35L, provided that you balance it to get rid of the upper rpm vibrations.



The EJ22 will always be inferior in potential when it comes to making power compared to the EJ25. That's not even a question. It's a simple matter of physics. There is no replacement for displacement. However, let's break down the mechanics to the EJ22 block, starting with the thicker liners. People have gone up to 27psi on these without so much as a cylinder wave or crack. The iron liners also hold up well in stock form.



The pistons are the next in line. Though not forged, they are strong enough that they can withstand 23-25psi with the correct tuning, air density, fueling, turbo, and heads. Many people have extracted as much as 380whp out of a stock EJ22E block before they lost ringlands. That being said, it's time to move onto the rods and crank, both of which are forged from the factory.

380hp is not enough to consider ej22 oem pistons strong, and by then you are already negating them being strong when you need forged to go above that level with 2.2l of displacement. We've all seen oem 2006 "weak" ringland pistons make close to 600hp and do 9s so how are you comparing pushing the limits to make "380hp" to obviously much higher potential on oem ringland stis?


have held up to 380whp without even a light bend. I've never seen a broken EJ22 crank, but I have seen broken EJ22 rods from lack of maintenance. Efficiency wise, since the crank and pistons are smaller, shorter stroked, and lighter than those in the EJ25, the efficiency as a hybrid for the EJ22 is greater, since there isn't as much internal inertia/resistance to spin up.

I have never seen a broken oem ej crank (excluding their diesels and flat sixes) due to high power levels. Are you hinting that the ej22 cranks are stronger than any of the other cranks produced from 92-01?

This all being said, the EJ25 in general is both inferior and superior to the EJ22 in many different ways. Build them up and sleeve them, and the EJ25 takes home the gold. But as a wild street engine, the EJ22 destroys it in reliability. It's a literal legend because all known EJ22 variants are known to be strong, as they can generally withstand more than the EJ25.

So what you are saying is that there are thousands of ej22 "wild" street cars roaming and waiting to take out "wild" ej25 street cars because they are more reliable? What? If this were the case, why not continue to produce the ej22 in turbo form in the sti cars instead of jumping to 2.5? I mean, if they are that tough with a small displacement, surely they could handle the abuse of all of the "wild" street builds and satisfy the need for power easily. Why aren't there hoards of them out roaming the streets hunting for victims?



They were phased out because Subaru had to comply with emissions regulations here, and the EJ22 just couldn't keep up with the demands, just barely being able to qualify. It is an amazing engine, with lethargic cylinder heads that much be replaced with something possessing a head cc of greater than 44 if one is to begin to see an example of the potential this engine is capable of unlocking.

What does being a 2.2 have to do with emissions? The 2.5 complies no problem. Why not continue to produce the 2.2 shortblock with the newer design heads and engine management if the 2.2 is so stout and surely more reliable?



Combination of years of notes, experience, and a LOT of financial mistakes for the sake of information gathering on these EJ's for the EJ22 hybrid builds.

I can agree on this, it has also taken me years as well to learn from spending cash but the knowledge gained is priceless. :)


EJ20H 9:1 Cam Specs:
Intake opens 8° BDTC.
Intake closes 52° ABDC.
Exhaust opens 52° BBDC.
Exhaust closes 8° ATDC.

EJ20H 8.5:1 Cam Specs:
Intake opens 10° BDTC.
Intake closes 50° ABDC.
Exhaust opens 52° BBDC.
Exhaust closes 2° ATDC.

EJ20G STi Cam Specs:
Intake opens 6° BDTC.
Intake closes 56° ABDC.
Exhaust opens 55° BBDC.
Exhaust closes 11° ATDC.
/\ Thanks. Do you have the ej20k non sti specs? Those are the cams and heads I run with my 257 block (modded combustion chambers)
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #30 · (Edited)
The block I received had very little wear, I split it and replaced bearings and o rings anyway to verify internal condition. You can't say wear you've seen is due to poor design if you don't know history of the engine. How are you determining "nearly 3-4x" wear?
I examined many engines myself, along with many others who documented their own in the past. I'd be littering this post with pictures of proof had my hard drive with the EJ info not fried.

"Our octane" Here it comes again. Please explain how US 93 octane is inferior to the Japanese premium. I've seen this before and watched it get shot down in a few threads.
Japanese Premium is around 97-98 octane. Ours is 93, tops, around here, anyway. Not only is that a crazy high octane, the Japanese ECU's are primitively programmed around that and their colder, more dense atmosphere. There is a massive difference between 93 and 98. Japan's lowest is 90 while ours is 87. That should give a clue to the kind of monsters the Japanese STi ECU's are. However, this is just programming. Tuning with a standalone gets around those limits here.

So now you're saying the ks are puking their guts out because of fuel and the gs didn't because they were just too strong and because of programming as well in the same paragraph. Please explain how strength of internals that you claim has anything to do with spun bearings. Even the closed deck versions are not going to have a better chance at preventing spun bearings.
Bearings = internal components that receive the maximum amount of stress during the combustion cycles. The earlier G's had bearings that were much stronger than that of the late K/G's, so they lasted longer than their "improved" counterparts. A full rebuild with upgraded components will get a K to last as long as any older G. It got so bad with the K's blowing bearings that people would buy them simply for their heads.

So now luck is why mine and many others' have made it out by the skin of their teeth? Look at my thread about the engines dying and you will notice that more than 50% of the ks show alive to the number of gs... and this is with the gs being produced for more than twice the number of years. This would indicate (if numbers were exact, but they can't be because not a large enough sample has been taken...but still) that the ks are as reliable or possibly better.
There is no way that closed deck G's were unreliable compared to the K's. It has always been known and proven that the orginal EJ20G's were the strongest of all the Phase 1 EJ20 turbos to have ever been produced. Now the Common G's and H's blowing up, I would agree, since many of them had a 9:1 compression ratio, and were used during a time when most of the G's were considered to be the same when they were different. Better heads, the K has, but definitely not better reliability than the old school G's.

How are you figuring in all of the ecu variants used with the "k" longblocks? I ran the 1S and my engine came with a VF22 from the factory which even with the "tame" ecu the end product was not. I can assure you my engine made over the gentleman's agreement the entire time I owned it and is still in awesome shape "internally" as I've split the block out of curiosity.
Maybe at some point the engine was either rebuilt, or driven like a grandma, or even owned by a passive Japanese grandpa/grandma and cared for meticulously? I've come across many EJ20R's with 40k on it that looked like hell on the inside, but then encountered some that were barely even driven. Guess it depends on the individual previous owner, but once our gas goes into those chambers, and the engines driven hard, all bets are off. Wear will accelerate.

There were obviously way more gs produced which contributes to their in service average being higher, and the fact they were originally tuned with less power contributes to their reliability overall. I sincerely doubt if you put an open deck g with an open deck k and tune them to the same hp levels and maintain them both equally, that one will be any more or less reliable. I've never seen any solid proof of this and doubt there ever will be, just a lot of conjecture and vague theory.
Common G's, along with H/K/R/D aren't as reliable as the older G's. It's all in the rod and crank bearings, really. Whatever Subaru did to those bearings from 94-98, it wasn't nearly as helpful as from 1989-1994. That much is certain. Not a speculation, just something people discovered a long while back. Now, what I HAVE discovered among those many things is that the bearings and rods from the old school G's and the EJ22E/T V1's are exactly the same, and so are those in the V2 EJ22E's. The JDM line in the mid 90's were all different.

Again with the "weaker internals", please explain how the g crank, rods, and pistons were stronger than the k? I'm talking about the non forged piston versions of course, to make things easier.
Tougher, more dense material, and tougher rod bearings used in the old G line made this possible. And Subaru used that same metallurgy in the other variants of the early EJ's as well. The late variants were not as reliable. Closed deck or not, everything EJ before 95 was awesome. Once 95 hit, reliability took a dive.

380hp is not enough to consider ej22 oem pistons strong, and by then you are already negating them being strong when you need forged to go above that level with 2.2l of displacement. We've all seen oem 2006 "weak" ringland pistons make close to 600hp and do 9s so how are you comparing pushing the limits to make "380hp" to obviously much higher potential on oem ringland stis?
I'm not basing this off of e85, I've seen their strength from 93 octane. e85 is a smokescreen. It makes everything seem stronger than they actually are when producing power. 380whp is quite powerful when you add 25% to it, which is what those EJ22 hybrids would be producing at the crank. Remember what I said before. In terms of power potential, the EJ25 will always win but when it comes to revolutions, response, and inherent toughness, the EJ22 destroys it by a large margin.

I have never seen a broken oem ej crank (excluding their diesels and flat sixes) due to high power levels. Are you hinting that the ej22 cranks are stronger than any of the other cranks produced from 92-01?
The cranks were a little more dense in the 90's, and so were the cylinder cases. The cases were more perfected after 1999, though. I've seen a broken EJ25 crank (though it was a 25D crank) and I've also seen more rods to come from the 2.5's than I would have ever wanted to. No EJ22 will go throw rods nearly half as dramatically as an EJ25 will. You're even more like to see them oval out their cylinders than crack walls like the EJ25's do.

So what you are saying is that there are thousands of ej22 "wild" street cars roaming and waiting to take out "wild" ej25 street cars because they are more reliable? What? If this were the case, why not continue to produce the ej22 in turbo form in the sti cars instead of jumping to 2.5? I mean, if they are that tough with a small displacement, surely they could handle the abuse of all of the "wild" street builds and satisfy the need for power easily. Why aren't there hoards of them out roaming the streets hunting for victims?
Thousands? I'd be surprised if there were that many. The Subaru community still seems a bit slow on the pickup of how much awesome potential the EJ22's have. All they know is the EJ22T, which is a narrow minded way to approach the EJ22. Subaru didn't know of the EJ22 potential back then.

Remember, in base form, the EJ22 is practically a truck engine built for simple moseying around, with some badass engine block internals, considering that most EJ22's were made when Subaru was shoving reliable parts in all of their engines. It wasn't until the EJ22G came out that people started to really pay attention to the EJ22, and began to build them in hybrid form.

What does being a 2.2 have to do with emissions? The 2.5 complies no problem. Why not continue to produce the 2.2 shortblock with the newer design heads and engine management if the 2.2 is so stout and surely more reliable?
It's the difficulty the EJ engines have always had, because in hindsight, they're constructed of an inefficient engine design. Inline 4's are and will always be better at producing power and controlled emissions than any boxer in existence, Subaru knows this, and has made many changes to the EJ22 in the 11 years it was in production to try and improve it.

What brought the EJ25 into existence in the first place were customers crying out for more TQ. When Subaru created it, they didn't expect too much out of it, and the first EJ25D was pretty much an overbored EJ20G. It performed better than expected, and they made several changes to its design over a decade. So many changes and improvements that it was able to live up to emissions expectations during a time when the EJ22 was struggling.

Reliability means nothing in the eyes of our government. Emissions standards are everything, and if Subaru wanted to keep playing the game that Malcolm Bricklin started, they would have to make a choice. They chose the EJ25 to succeed the EJ22, and placed it in every car from 2002 onward. Is it possible that a factory built hybrid EJ22 could have passed emissions? Absolutely, as mine did. However, it was no longer what the people wanted for a powerplant.
 

·
Registered
The subaru of many acronyms
Joined
·
1,919 Posts
It seems like your taking this a bit far as to believe the ej22 shortblock is some holy grail and all other blocks fall drastically short of the ej22s capability. Is it a good platform for a cheap throw together project? sure. Is it honestly overall better than a stock turbo ej20/25? not really. Each engine has its benefits and drawbacks, there is no perfect engine from the factory (ignoring group N factory engines etc). Whats the best stock block for turbo? The one the best fits your needs, ultimately power is going to be limited to the cylinder head and turbo assuming equals among other accessories. The ej20 will have a wider power band up top but be slightly laggy, the ej25 will spool up quicker with better midrange torque, but fall off up top, and the ej22 will be somewhere in the middle. With knock, abuse, excess heat, any critical component piston connecting rod bearings crank etc is susceptible to damage. Would I ever invest any money into building an ej22 open deck block? no. You can sleeve 2.5's for big power, or just buy closed deck ej20g, ver10+ ej207, balance the components and slap some t20 heads with aggressive cams and rev that sunabitch out. Or if your after ultimate power you sleeve a eg33 and go from there.
 

·
needs more cowbell
95L, 07 2.5i, R6
Joined
·
9,315 Posts
double post, sorry
 

·
needs more cowbell
95L, 07 2.5i, R6
Joined
·
9,315 Posts
Stronger bearings? Denser metals? I give up, lol


I examined many engines myself, along with many others who documented their own in the past. I'd be littering this post with pictures of proof had my hard drive with the EJ info not fried.



Japanese Premium is around 97-98 octane. Ours is 93, tops, around here, anyway. Not only is that a crazy high octane, the Japanese ECU's are primitively programmed around that and their colder, more dense atmosphere. There is a massive difference between 93 and 98. Japan's lowest is 90 while ours is 87. That should give a clue to the kind of monsters the Japanese STi ECU's are. However, this is just programming. Tuning with a standalone gets around those limits here.



Bearings = internal components that receive the maximum amount of stress during the combustion cycles. The earlier G's had bearings that were much stronger than that of the late K/G's, so they lasted longer than their "improved" counterparts. A full rebuild with upgraded components will get a K to last as long as any older G. It got so bad with the K's blowing bearings that people would buy them simply for their heads.



There is no way that closed deck G's were unreliable compared to the K's. It has always been known and proven that the orginal EJ20G's were the strongest of all the Phase 1 EJ20 turbos to have ever been produced. Now the Common G's and H's blowing up, I would agree, since many of them had a 9:1 compression ratio, and were used during a time when most of the G's were considered to be the same when they were different. Better heads, the K has, but definitely not better reliability than the old school G's.



Maybe at some point the engine was either rebuilt, or driven like a grandma, or even owned by a passive Japanese grandpa/grandma and cared for meticulously? I've come across many EJ20R's with 40k on it that looked like hell on the inside, but then encountered some that were barely even driven. Guess it depends on the individual previous owner, but once our gas goes into those chambers, and the engines driven hard, all bets are off. Wear will accelerate.



Common G's, along with H/K/R/D aren't as reliable as the older G's. It's all in the rod and crank bearings, really. Whatever Subaru did to those bearings from 94-98, it wasn't nearly as helpful as from 1989-1994. That much is certain. Not a speculation, just something people discovered a long while back. Now, what I HAVE discovered among those many things is that the bearings and rods from the old school G's and the EJ22E/T V1's are exactly the same, and so are those in the V2 EJ22E's. The JDM line in the mid 90's were all different.



Tougher, more dense material, and tougher rod bearings used in the old G line made this possible. And Subaru used that same metallurgy in the other variants of the early EJ's as well. The late variants were not as reliable. Closed deck or not, everything EJ before 95 was awesome. Once 95 hit, reliability took a dive.



I'm not basing this off of e85, I've seen their strength from 93 octane. e85 is a smokescreen. It makes everything seem stronger than they actually are when producing power. 380whp is quite powerful when you add 25% to it, which is what those EJ22 hybrids would be producing at the crank. Remember what I said before. In terms of power potential, the EJ25 will always win but when it comes to revolutions, response, and inherent toughness, the EJ22 destroys it by a large margin.



The cranks were a little more dense in the 90's, and so were the cylinder cases. The cases were more perfected after 1999, though. I've seen a broken EJ25 crank (though it was a 25D crank) and I've also seen more rods to come from the 2.5's than I would have ever wanted to. No EJ22 will go throw rods nearly half as dramatically as an EJ25 will. You're even more like to see them oval out their cylinders than crack walls like the EJ25's do.



Thousands? I'd be surprised if there were that many. The Subaru community still seems a bit slow on the pickup of how much awesome potential the EJ22's have. All they know is the EJ22T, which is a narrow minded way to approach the EJ22. Subaru didn't know of the EJ22 potential back then.

Remember, in base form, the EJ22 is practically a truck engine built for simple moseying around, with some badass engine block internals, considering that most EJ22's were made when Subaru was shoving reliable parts in all of their engines. It wasn't until the EJ22G came out that people started to really pay attention to the EJ22, and began to build them in hybrid form.



It's the difficulty the EJ engines have always had, because in hindsight, they're constructed of an inefficient engine design. Inline 4's are and will always be better at producing power and controlled emissions than any boxer in existence, Subaru knows this, and has made many changes to the EJ22 in the 11 years it was in production to try and improve it.

What brought the EJ25 into existence in the first place were customers crying out for more TQ. When Subaru created it, they didn't expect too much out of it, and the first EJ25D was pretty much an overbored EJ20G. It performed better than expected, and they made several changes to its design over a decade. So many changes and improvements that it was able to live up to emissions expectations during a time when the EJ22 was struggling.

Reliability means nothing in the eyes of our government. Emissions standards are everything, and if Subaru wanted to keep playing the game that Malcolm Bricklin started, they would have to make a choice. They chose the EJ25 to succeed the EJ22, and placed it in every car from 2002 onward. Is it possible that a factory built hybrid EJ22 could have passed emissions? Absolutely, as mine did. However, it was no longer what the people wanted for a powerplant.
 

·
Registered
The subaru of many acronyms
Joined
·
1,919 Posts
Oh and differences in RON vs AKI for computing octane means 98 octane in japan is actually a touch over 93 octane u.s. so the octane thing doesn't really hold true unless your 91 octane in your state or your cali/az water gas.
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #35 ·
It seems like your taking this a bit far as to believe the ej22 shortblock is some holy grail and all other blocks fall drastically short of the ej22s capability. Is it a good platform for a cheap throw together project? sure. Is it honestly overall better than a stock turbo ej20/25? not really.
The EJ22T has often been called "the holy grail" of Subaru engines, but this was during a time when the older G's were pricey and hard to get ahold of.

EJ20's fall short of the EJ22's in certain areas, as do the EJ25's. As I have mentioned before, they are the perfect midpoint between the 2 engines.

As a turbo engine, the EJ22's are amazing. Any EJ22 builder here will tell you this. It has more power potential than an EJ20 and is more reliable than an EJ25. What's not to like about it?

Would I ever invest any money into building an ej22 open deck block? no. You can sleeve 2.5's for big power, or just buy closed deck ej20g, ver10+ ej207, balance the components and slap some t20 heads with aggressive cams and rev that sunabitch out. Or if your after ultimate power you sleeve a eg33 and go from there.
I would invest money into building an EJ22 faster than I would ever waste money on an EJ25, especially if I wanted a capable and reliable track car that comes with a measure of efficiency. Open deck block EJ22's are tougher than semi-closed EJ25's. Strike one. Open deck EJ22's are far more plentiful and cheaper than semi-closed EJ25's. Strike 2. Open deck EJ25's are more reliable than semi-closed EJ25's. Strike 3.

With that being said, If I had an EJ22T block, I'd get it cut for 5th position main, drop a 257 crank into it, install forged rods and pistons, and drop on some hot 207 heads. 2.3L monster a 2L would have a hard time keeping pace with, as it'll boost up faster and sooner than any stock EJ25 turbo. Remember, the EJ22 has an EJ20 crank, so the revs are there to be found.

Now, the only way you could get a 2.5L to surpass an EJ22 hybrid IS to sleeve it and do all that extra ish, but I like simplicity, and it doesn't get any more simple than a $300 low-mileage junkyard engine block that has a greater potential for power than any EJ20 turbo, and would last much longer than any stock EJ25. The EJ22 hybrid is the poor man's way of getting around those expensive turbo EJ25 engines, and it works savagely.
 

·
Registered
The subaru of many acronyms
Joined
·
1,919 Posts
Your $300 low budget monster requires custom connecting rods and pistons $$$$, and special machining to run the crank you want. Plus bore matching on the cylinder head. Low budget monster? no.
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #37 ·
Stronger bearings? Denser metals? I give up, lol
There is info substantiating this over 10yrs by other people on many different forums, NASIOC being one of the biggest ones on that period of time. A search there would reveal tons of information. Ultimate Subaru and Legacy Central being other places where key players have entered useful data.

Oh and differences in RON vs AKI for computing octane means 98 octane in japan is actually a touch over 93 octane u.s. so the octane thing doesn't really hold true unless your 91 octane in your state or your cali/az water gas.
You're also not factoring the amount of fuel and the programming the aggressive JDM ECU's have over their USDM counterparts. And I did the math. The AFAIK of the Premium in Japan is closer to 94-95, which means I correct myself earlier statement. It's still not safe for 93 to be run too hard in the upper aggressive ECU's.
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #38 ·
Your $300 low budget monster requires custom connecting rods and pistons $$$$, and special machining to run the crank you want. Plus bore matching on the cylinder head. Low budget monster? no.
It'd still be cheaper than setting up an EJ25. LOL! And I don't have to bore match with some cylinder heads, like the 04-07 D25's and the 251 heads. Hell, if I wanted some low-end, I wouldn't have to bore match at all with any clover heads, at the clovers themselves are almost matched to the 97mm bore (really, they're matched up with the EJ20 bore exactly.) There are many tricks to the EJ22 hybrid, as the block itself has many EJ20 and EJ25 heads which can be chosen for it.
 

·
Registered
The subaru of many acronyms
Joined
·
1,919 Posts
If you think you can successfully run a pentroof 99.5mm bore ej25 head on a 97mm ej22 block, your gonna have a bad time. The 25d heads are 20k head castings, but for anything more than a throw me together block its not worth it, there not as efficient of a design, and theyre is not as much aftermarket support. For an unmodified block as a building platform you cannot surpass the ver10+ ej207 or ej20g RA block.
 

·
Registered
1998 Subaru Legacy GT Limited
Joined
·
236 Posts
Discussion Starter · #40 · (Edited)
If you think you can successfully run a pentroof 99.5mm bore ej25 head on a 97mm ej22 block, your gonna have a bad time.
I explained this same exact thing earlier in the thread. The 1996 25D heads are no good for this application. Those heads are better left on EJ25 blocks.

The 25d heads are 20k head castings, but for anything more than a throw me together block its not worth it, there not as efficient of a design, and theyre is not as much aftermarket support.
This is false, because not only do these heads allow sufficient air to come in as is, they have the potential to rival ported 205 heads. De-shroud the chambers and you have a good all-around combustion chamber for performance, not to mention there are different sets of stock cams that will just drop right in, like the STi 20K cams. Example, in N/A format, I was able to achieve almost 36hwy mpg in the 2nd EJ22D I ever built. They breathe well. These heads have aftermarket support and great potential.

For an unmodified block as a building platform you cannot surpass the ver10+ ej207 or ej20g RA block.
The EJ22T block is pretty much a bored early EJ20G block, which is the same used in the early RA's. These blocks, though probably not as strong as the early RA's, have not yet been proven weaker.
 
21 - 40 of 80 Posts
Top